hardware compatibility and patent infringement 
Author Message
 hardware compatibility and patent infringement

Hi,

I need some "light" on "hardware compatibility and patent infringement".
Since Transmeta presented his Crusoe chip I've had a few doubts about
this issue. I would much appreciate any information on the following
points you could give:

1. Is Transmeta's SOFTWARE approach to the manipulation of x86
instructions set what avoids the violation of patents' rights? If that's
right, I understand that patent infringement is only possible when there
is a hardware implementation.

2. In general (for any kind of processor), What's the way to design a
hardware device compatible with the instruction set of another processor
without infringement of patents?

3. If someone implements a processor compatible with the instruction set
of another processor but without copying anyting else from that original
processor    
a part from the original instruction set, will that be patent
infringement?

Although I've been looking quite thoroughly on the web for information
on patents for electronics design (the technical side of the subject)
I've had little success. Can anyone, please, point me to a good source
of information on this subject.

Thank you all very much for your time.

A. Petit



Sat, 01 Mar 2003 22:58:43 GMT  
 hardware compatibility and patent infringement

Quote:

> Hi,

> I need some "light" on "hardware compatibility and patent infringement".
> Since Transmeta presented his Crusoe chip I've had a few doubts about
> this issue. I would much appreciate any information on the following
> points you could give:

> 1. Is Transmeta's SOFTWARE approach to the manipulation of x86
> instructions set what avoids the violation of patents' rights? If that's
> right, I understand that patent infringement is only possible when there
> is a hardware implementation.

> 2. In general (for any kind of processor), What's the way to design a
> hardware device compatible with the instruction set of another processor
> without infringement of patents?

> 3. If someone implements a processor compatible with the instruction set
> of another processor but without copying anyting else from that original
> processor
> a part from the original instruction set, will that be patent
> infringement?

> Although I've been looking quite thoroughly on the web for information
> on patents for electronics design (the technical side of the subject)
> I've had little success. Can anyone, please, point me to a good source
> of information on this subject.

> Thank you all very much for your time.

> A. Petit

My impression is that you can't patent an instruction set.  Then again,
a lot of patents have been granted that are plain stupid.
In your favor, there are several Intel clone chips, that Intel would love
to shut down if they could.

The implimentation of an instruction set probably can be patented.



Sat, 01 Mar 2003 16:15:35 GMT  
 hardware compatibility and patent infringement

Quote:

> 1. Is Transmeta's SOFTWARE approach to the manipulation of x86
> instructions set what avoids the violation of patents' rights? If that's
> right, I understand that patent infringement is only possible when there
> is a hardware implementation.

This is not correct...  At least not according to MIPS's lawyers.  MIPS is
currently suing Lexra for doing software emulation of some patented
instructions (LWL, LWR, SWL, and SWR to be exact).  Of course
Lexra's lawyers would disagree.

Quote:
> 2. In general (for any kind of processor), What's the way to design a
> hardware device compatible with the instruction set of another processor
> without infringement of patents?

If an instruction is patented, and that is possible, then as long as you do
that instruction you are a target for a lawsuit.  There is no way around
that,
other than waiting for the patent to expire or licensing the patent.

Quote:
> 3. If someone implements a processor compatible with the instruction set
> of another processor but without copying anyting else from that original
> processor a part from the original instruction set, will that be patent
> infringement?

If you do what is described in the patent then you're infringing.  If it is
possible
to emulate an instruction without doing what is described in the patent then
you're
OK.

In the case of MIPS vs. Lexra it is impossible emulate the instruction
without
doing what's in the patent since the patent describes the instructions
themselves.
What is at issue in MIPS vs. Lexra is the software emulation of the
hardware.

A possible workaround for the MIPS vs Lexra dispute is for Lexra to modify
the C/C++ compiler to never generate the instructions in dispute and emulate
it
"in a typical fashion that's been done for 20+  years".  By never doing
those
instructions they cannot be infringing.  Of course, this limits the
"compatibility"
of their CPU.

David Kessner



Sun, 02 Mar 2003 00:44:32 GMT  
 hardware compatibility and patent infringement
Quote:

> This is not correct...  At least not according to MIPS's lawyers.  MIPS is
> currently suing Lexra for doing software emulation of some patented
> instructions (LWL, LWR, SWL, and SWR to be exact).  Of course
> Lexra's lawyers would disagree.

1) What do these instructions do? (They look like 'word shift' type
operations....)

2) Did they patent the instruction menomic or what the instruction did? If
its the latter then a patent has been issued for an algorithm. I do hope its
not a commonly employed one - I can't think of any operations at the
register level that haven't been implemented in some machine or another over
the last five decades.

2a) Then there's the issue of instruction emulation by microcode. This
sounds fancy and modern but the technique has been around for decades - its
the logical way to implement complex instructions on a processor. Emulating
other people's instructions isn't new. I first became aware of a machine
that could do this in the mid 70s - it was offered as a way of running
software from a new range of computers and the company's legacy systems.

2b) I think everyone agrees that the patent office needs reinventing. It
currently does more damage than good in our business, IMHO.

3) Intel didn't patent their instruction set. They copyrighted it. That's
why NEC had to come up with creative alternative names for the same
instructions in their 'V' series. AMD and the like were OK due to some
second sourcing agreement.

4) I've got a real 'thing' about people who patent bullshit things,
especially if they really believe that they invented this stuff. It takes a
special combination of arrogance and ignorance to believe that you have
truly invented something basic in our trade. They need to humble up, learn a
bit, and maybe work hard enough to actually invent something new.



Sun, 02 Mar 2003 11:33:52 GMT  
 hardware compatibility and patent infringement

[snip]

 Martin> 4) I've got a real 'thing' about people who patent bullshit
 Martin> things, especially if they really believe that they invented
 Martin> this stuff. It takes a special combination of arrogance and
 Martin> ignorance to believe that you have truly invented something
 Martin> basic in our trade. They need to humble up, learn a bit, and
 Martin> maybe work hard enough to actually invent something new.

[facetiousness on]

I have a patent pending on the use of the * character in `C' code.
And may I say I am brilliant for deciding to finally use this
character that is defined in the `C' standard, IMHO. This explains why
C++ has a novel use of the "&" character.  Please switch now or be
sued.

[facetiousness off]

Bill



Sun, 02 Mar 2003 12:53:19 GMT  
 hardware compatibility and patent infringement

Quote:
> > 1. Is Transmeta's SOFTWARE approach to the manipulation of x86
> > instructions set what avoids the violation of patents' rights? If
that's
> > right, I understand that patent infringement is only possible when
there
> > is a hardware implementation.

> This is not correct...  At least not according to MIPS's lawyers.

Completely different issue.  The instruction set is not patented, nor can
it be.

Quote:
> MIPS is
> currently suing Lexra for doing software emulation of some patented
> instructions (LWL, LWR, SWL, and SWR to be exact).  Of course
> Lexra's lawyers would disagree.

Here is the patent:

http://patent.womplex.ibm.com/details?&pn=US04814976__

I have written code to do the same thing for over 20 years, and I know I
have seen other code before mine that did the same thing.  I even did a
microcode implementation similar to this in a 29203C some 20 or so years
ago for a vision processor.  There is sure to be plenty of prior art to
this...  I downloaded a copy of the patent in its' entirety, and perhaps
there is something really patentable here, but from my initial look, it
seems like this won't hold up.  Just my opinion...

Here are the 14 'claims':

What is claimed is:
1. In a reduced instruction set computer with a memory holding m-bit words
separated by word boundaries, a device for retrieving an unaligned
reference from said memory comprising:
a. a general register;
b. means for retrieving a first word containing a first portion of said
unaligned reference in response to a nth instruction and a second word
containing a second portion of said unaligned reference from said memory in
response to an (n+k)th instruction;
c. shifting means for shifting said first portion to a first position and
second portion to a second position; and
d. combining means for combining said first and second portions in said
general register, wherein k and n are positive integers.
2. In reduced instruction set computer, a device for storing an unaligned
reference into a memory with m-bit locations comprising:
shifting means for shifting said unaligned reference in a first direction
in response to an nth instruction and in a second direction in response to
(n+k)th instruction, said means generating sequentially a first and second
portion each having less than m-bits; and
means for storing said first and second portions sequentially into said
memory, wherein k and n are positive integers.
3. In a reduced instruction set computer with a memory for holding m-bit
words, a device for loading a first unaligned reference having first and
second portions of less than m-bits, said first portion being stored into a
first section of said memory and said second portion being stored into a
second section of said memory, and for storing a second aligned reference
into said first and said second sections, comprising:
a shift/merge unit having first and second inputs and being provided to
shift first data bytes received from said first input, said first input
being coupled to said memory unit to receive said first and second portions
sequentially, and merge said first data bytes with second data bytes from
said second input to form an m-bit word;
a first latch means for storing said first and second data bytes, said
latch having an output coupled to said second input;
an m-bit general register coupled to said first latch means and provided
for holding selectively one of said first or second unaligned references;
a second latch means coupled to said register for storing said second
unaligned reference;
shifting means for shifting said second unaligned references; and
output means for storing said second unaligned reference after shifting by
said shifting means into said memory.
4. The device of claim 3 wherein said shift/merge unit shifts bytes
received from said memory in a first direction in response to a first load
instruction, and in a second direction in response to a second load
instruction.
5. The device of claim 3 wherein said shifting means shifts bytes received
from said second latch means in a first direction in response to a first
store instruction, and in a second direction in response to a second store
instruction.
6. The device of claim 3 further comprising a bypass multiplexer for
selectively coupling to said second input one of the outputs of said first
and second latching means.
7. The device of claim 4 wherein said first and second load instructions
are at least partially overlapped.
8. The device of claim 5 wherein said first and second store instructions
are at least partially overlapped.
9. A method of loading an m-bit unaligned reference from a memory, said
memory holding m-bit words separated by word boundaries, said m-bit
unaligned reference being divided into a first portion and a second portion
by a word boundary, comprising the steps of:
a. retrieving a first word from said memory containing said first portion
during an (nth) instruction;
b. shifting said first portion to a first position;
c. retrieving a second word containing said second portion during an
(n+k)th instruction;
d. shifting said second portion to a second position; and
e. merging said first and second portions;
wherein said k and n are positive integers and wherein said first and
second portions have less than m bits.
10. The method of claim 9 wherein said first and second positions are
defined by said nth and (n+k)th instruction respectively.
11. The method of claim 9 wherein said nth and (n+k)th instructions are
overlapped.
12. A method of storing an unaligned reference into a computer memory, said
computer memory holding m-bit locations separated by word boundaries,
comprising the steps of:
a. shifting a first portion of said reference to a first portion;
b. storing said first portion in one location within an nth instruction;
c. shifting a second portion of second portion of said reference to a
second position; and
d. storing said second portion to a second location within an (n+k)th
instruction, wherein n and k are positive integers and wherein said first
and second portions have less than m bits.
13. The method of claim 12 wherein said first and second position are
defined by said nth and (n+k)th instruction respectively.
14. The method of claim 12 wherein said nth and (n+k)th instructions are
overlapped.



Mon, 03 Mar 2003 10:17:14 GMT  
 hardware compatibility and patent infringement

Quote:

> > This is not correct...  At least not according to MIPS's lawyers.  MIPS is
> > currently suing Lexra for doing software emulation of some patented
> > instructions (LWL, LWR, SWL, and SWR to be exact).  Of course
> > Lexra's lawyers would disagree.

> 1) What do these instructions do? (They look like 'word shift' type
> operations....)

They are "Load Word Left/Right" and "Store Word Left/Right".  They are
used when loading & storing words that are not aligned on a word boundary.

Quote:
> 2) Did they patent the instruction menomic or what the instruction did? If
> its the latter then a patent has been issued for an algorithm. I do hope its
> not a commonly employed one - I can't think of any operations at the
> register level that haven't been implemented in some machine or another over
> the last five decades.

They patented what the instruction did.  One could argue that they didn't
patent an algorithm-- only a way of doing an algorithm.  These instructions
are basically do a load-shift-and-or.  We could do these instructions using
those standard 4 instructions, but the patent covers doing them in a single
instruction.

Quote:
> 2b) I think everyone agrees that the patent office needs reinventing. It
> currently does more damage than good in our business, IMHO.

Amen to that!

Quote:
> 4) I've got a real 'thing' about people who patent bullshit things,
> especially if they really believe that they invented this stuff. It takes a
> special combination of arrogance and ignorance to believe that you have
> truly invented something basic in our trade. They need to humble up, learn a
> bit, and maybe work hard enough to actually invent something new.

I think I'll patent the "Look and Feed Like S&*T".  I would just love to
sue Microsoft, claiming that their software violates the "Look and Feed Like
S&*T"
patent.  Too bad about that prior art thing...

David Kessner



Mon, 03 Mar 2003 03:00:00 GMT  
 hardware compatibility and patent infringement

Quote:

> > > 1. Is Transmeta's SOFTWARE approach to the manipulation of x86
> > > instructions set what avoids the violation of patents' rights? If
> that's
> > > right, I understand that patent infringement is only possible when
> there
> > > is a hardware implementation.

> > This is not correct...  At least not according to MIPS's lawyers.

> Completely different issue.  The instruction set is not patented, nor can
> it be.

Apparently it can be!  Of course, we have to really look at what was
patented...

What was patented is the combination of several common things into
four instructions.  Not the instruction itself, its neumonic, or the opcode
encoding.
The combination of operations into some simple instructions is what's patented.

And this, I believe, can be patented (as much as I disagree with patents).

Quote:
> > MIPS is
> > currently suing Lexra for doing software emulation of some patented
> > instructions (LWL, LWR, SWL, and SWR to be exact).  Of course
> > Lexra's lawyers would disagree.

> Here is the patent:

> http://www.*-*-*.com/

> I have written code to do the same thing for over 20 years, and I know I
> have seen other code before mine that did the same thing.

But that doesn't matter...  We've all done code that that.  What's patented
is that this has been put into some instructions specifically for this purpose.

Where Lexra's getting into trouble is that they are doing a software solution
to
hardware emulation.  Essentially, Lexra is still executing the instructions
(but
in software) and MIPS is saying that what matters is that they are being
executed at all (SW or HW).

Quote:
> I even did a
> microcode implementation similar to this in a 29203C some 20 or so years
> ago for a vision processor.

I'm not sure if that applies.  The patent refers to a RISC CPU (i.e., no
microcode).

Quote:
> There is sure to be plenty of prior art to
> this...  I downloaded a copy of the patent in its' entirety, and perhaps
> there is something really patentable here, but from my initial look, it
> seems like this won't hold up.  Just my opinion...

I agree on the prior art thing, and with your general opinion.  But...

Our patent system is broken.  If I were on trial for {*filter*}, my punishment
would
come after the judge renders his verdict.  But if I were on trial for patent
infringement,
my punishment would come during the trial, in the form of trial expenses, not
once the
judge gives his verdict.

This is a double whammy...  Not only does the patent office give frivolous
patents, but
there is no presumed innocence.  The innocently accused will always loose,
regardless
of what the judge says.

The only protection for this type of lawsuit is to counter sue for some other
patent
infringement.  The two sides lock in a stalemate and are forced to settle (this
is called
a "cross licensing agreement").  This is one huge reason why looser companies
are bought
out-- since their patent portfolio can be more valuable than the actual
company.  Buy
a small company for 10 million, but save 20 million in legal expenses...

I guess my point is that the small guy has to be extra careful about this
patent
stuff since it's really easy to get stepped on and squished into oblivion (been
there,
done that).

David Kessner



Mon, 03 Mar 2003 03:00:00 GMT  
 hardware compatibility and patent infringement

Quote:

> 2) Did they patent the instruction menomic or what the instruction did? If
> its the latter then a patent has been issued for an algorithm. I do hope its
> not a commonly employed one - I can't think of any operations at the
> register level that haven't been implemented in some machine or another over
> the last five decades.

It not legal, but it should be legal to patent algorithms assuming you actually
create a significantly new algorithm which is nontrivial to say the least.  A
new algorithm is a significant contribution to the world, and it would be
wonderful for more money to be spent on developing new algorithms.  Software
code on the other hand is simply the implementation of a set of algorithms and
should not be patentable any more than a short story, or in the case of
M$ product an library of books.  Software is properly covered by copy write
laws.  

There should be no threat of people patenting
existing algorithms, since unless a person can prove they invented something
first and started the patent process with reasonable dispatch, the patent
is invalid.  So I can't patent Quick Sort and A*, there goes that easy
trillion bucks.  

Muddy

Quote:

> 4) I've got a real 'thing' about people who patent bullshit things,
> especially if they really believe that they invented this stuff. It takes a
> special combination of arrogance and ignorance to believe that you have
> truly invented something basic in our trade. They need to humble up, learn a
> bit, and maybe work hard enough to actually invent something new.

Until Americans acknowledge that the Government needs truly qualified people
and that to get them will cost real money these problems will continue.  You
can't expected an agency forced to staff its offices with engineers that
are paid 20-30% less than private industry, and got C's in school to be
able to monitor the volume of patent applications they receive.


Mon, 03 Mar 2003 03:00:00 GMT  
 hardware compatibility and patent infringement

Quote:


> > > This is not correct...  At least not according to MIPS's lawyers.  MIPS is
> > > currently suing Lexra for doing software emulation of some patented
> > > instructions (LWL, LWR, SWL, and SWR to be exact).  Of course
> > > Lexra's lawyers would disagree.

> > 1) What do these instructions do? (They look like 'word shift' type
> > operations....)

> They are "Load Word Left/Right" and "Store Word Left/Right".  They are
> used when loading & storing words that are not aligned on a word boundary.

> > 2) Did they patent the instruction menomic or what the instruction did? If
> > its the latter then a patent has been issued for an algorithm. I do hope its
> > not a commonly employed one - I can't think of any operations at the
> > register level that haven't been implemented in some machine or another over
> > the last five decades.

> They patented what the instruction did.  One could argue that they didn't
> patent an algorithm-- only a way of doing an algorithm.  These instructions
> are basically do a load-shift-and-or.  We could do these instructions using
> those standard 4 instructions, but the patent covers doing them in a single
> instruction.

um... doesn't motorola have such an instruction in the 680x0 series?

David



Mon, 03 Mar 2003 03:00:00 GMT  
 hardware compatibility and patent infringement
Easy big fella... I am one of those ones who made some 'C's in college
but I turned out fairly decent. Dang, now I have a complex!

-Al Arduengo
Motorola SPS / Digital Audio Organization

Quote:

> Until Americans acknowledge that the Government needs truly qualified people
> and that to get them will cost real money these problems will continue.  You
> can't expected an agency forced to staff its offices with engineers that
> are paid 20-30% less than private industry, and got C's in school to be
> able to monitor the volume of patent applications they receive.



Mon, 03 Mar 2003 03:00:00 GMT  
 hardware compatibility and patent infringement

Quote:

> Easy big fella... I am one of those ones who made some 'C's in college
> but I turned out fairly decent. Dang, now I have a complex!

Well most programers program at the  C level... some even make it
to C++ :)
--
"We do not inherit our time on this planet from our parents...
 We borrow it from our children."
"Luna family of Octal Computers" http://www.jetnet.ab.ca/users/bfranchuk


Mon, 03 Mar 2003 03:00:00 GMT  
 hardware compatibility and patent infringement

Quote:

> Hi,

> I need some "light" on "hardware compatibility and patent infringement".
> Since Transmeta presented his Crusoe chip I've had a few doubts about
> this issue. I would much appreciate any information on the following
> points you could give:

> 1. Is Transmeta's SOFTWARE approach to the manipulation of x86
> instructions set what avoids the violation of patents' rights? If that's
> right, I understand that patent infringement is only possible when there
> is a hardware implementation.

> 2. In general (for any kind of processor), What's the way to design a
> hardware device compatible with the instruction set of another processor
> without infringement of patents?

> 3. If someone implements a processor compatible with the instruction set
> of another processor but without copying anyting else from that original
> processor
> a part from the original instruction set, will that be patent
> infringement?

> Although I've been looking quite thoroughly on the web for information
> on patents for electronics design (the technical side of the subject)
> I've had little success. Can anyone, please, point me to a good source
> of information on this subject.

> Thank you all very much for your time.

> A. Petit

My layman's take on this (though I did some following of the Intel et al CPU
wars some years back) is that Transmeta is somewhat protected from patent
violations by the fact it is using fab labs of companies that have patent
cross-licensing agreements with Intel, or have acquired them by purchasing
companies that had these.

I would not be surprised, however, if Transmeta sales start to eat into
Intel's sales, Intel's lawyers will file something. Even if Intel doesn't
win, it can really wreck havoc on a small company like Transmeta to have its
the bulk of its sales/marketing/managment and engineering talent sitting in
a courtroom defending its technology and its products rather than working
back at the company making it a sales success.

Sincerely,

 MarcW.



Mon, 03 Mar 2003 03:00:00 GMT  
 hardware compatibility and patent infringement

Quote:

> 4) I've got a real 'thing' about people who patent bullshit things,
> especially if they really believe that they invented this stuff. It takes a
> special combination of arrogance and ignorance to believe that you have
> truly invented something basic in our trade. They need to humble up, learn a
> bit, and maybe work hard enough to actually invent something new.

I don't think they care about inventing anything. They care about making
money. The patent/copyright law is a good vehicle for making a quick
buck. Today neither has anything to do with creating something, apart
from wealth, of course, for the IP owners and the lawyers.

Zoltan

--
+------------------------------------------------------------------+
| ** To reach me write to zoltan in the domain of bendor com au ** |
+--------------------------------+---------------------------------+
| Zoltan Kocsi                   |   I don't believe in miracles   |  
| Bendor Research Pty. Ltd.      |   but I rely on them.           |
+--------------------------------+---------------------------------+



Mon, 03 Mar 2003 03:00:00 GMT  
 hardware compatibility and patent infringement


Quote:
>My impression is that you can't patent an instruction set.  Then again,
>a lot of patents have been granted that are plain stupid.
>In your favor, there are several Intel clone chips, that Intel would love
>to shut down if they could.

        I believe Intel has some patents on the IA64 ISA, but then
there are some significant new and novel features on it.
--



Tue, 04 Mar 2003 09:02:48 GMT  
 
 [ 37 post ]  Go to page: [1] [2] [3]

 Relevant Pages 

1. hardware compatibility and patent infringement

2. VHDL/Verilog/Ada & software/hardware patent equivalence

3. VHDL/Verilog/Ada & software/hardware patent equivalence

4. GNU Smalltalk - ? Hardware compatibility

5. PC-Scheme (TI) and hardware compatibility

6. Hardware independant hardware access ...

7. Update Patent field

8. Open Standards vs Patents

9. WebBook Patent protection

10. Software patent question

11. PATENTED INVENTIONS WANTED!!

12. VHDL/CAE software patents from 1992/93

 

 
Powered by phpBB® Forum Software